Claude Skills
CollectionsCompareWorkflowsNominate
Sign inSign up
© 2026 Curated Agent Skills·Learn more about Agent Skills
Back to repository

synthesizing-pillars

verified

Use when evidence collection is complete for a pillar and need to extract actionable insights. Transforms raw evidence into structured synthesis with patterns and contradictions identified.

View on GitHub

Marketplace

synapti-marketplace

synaptiai/synapti-marketplace

Plugin

context-ledger

product-development

Repository
Verified Org

synaptiai/synapti-marketplace

plugins/context-ledger/skills/synthesizing-pillars/SKILL.md

Last Verified

February 4, 2026

Install Skill

Select agents to install to:

Scope:
npx add-skill https://github.com/synaptiai/synapti-marketplace/blob/main/plugins/context-ledger/skills/synthesizing-pillars/SKILL.md -a claude-code --skill synthesizing-pillars

Installation paths:

Claude
.claude/skills/synthesizing-pillars/
Powered by add-skill CLI

Instructions

# Pillar Synthesis

This skill transforms raw evidence objects into structured insights for a single research pillar.

## Prerequisites

- Evidence gate passed (≥5 evidence objects for this pillar)
- Evidence objects in `02-evidence/<pillar>/`

## Workflow

Use TodoWrite to track these mandatory steps:

<required>
1. Load all evidence objects for the pillar
2. Identify patterns and themes
3. Resolve or document contradictions
4. Extract key insights
5. Generate synthesis document
6. Link insights to evidence IDs
</required>

### Step 1: Load Evidence

Read all `EV-<pillar>-*.yaml` files from `02-evidence/<pillar>/`.

For each evidence object, extract:
- Claim
- Confidence
- Assumptions
- Tags

Build a working set of all claims.

### Step 2: Identify Patterns

Group evidence by theme:
- What claims cluster together?
- What topics have multiple evidence points?
- What themes emerge across sources?

**Pattern identification:**
```
Theme: Pricing
├── EV-market-pricing-smb-wtp (0.75) - SMB WTP $29
├── EV-market-pricing-enterprise-wtp (0.80) - Enterprise WTP $99
└── EV-competitors-pricing-benchmark (0.70) - Market avg $45

Insight: Pricing flexibility needed for multi-segment
```

### Step 3: Handle Contradictions

For contradictory evidence:

1. **Note the contradiction explicitly**
2. **Assess confidence delta** - Higher confidence wins if large gap
3. **Check recency** - More recent may supersede older
4. **Check authority** - More authoritative source wins
5. **If unresolved** - Document both, flag for decision-making

```markdown
### Contradiction: User Segment Priority

**EV-users-segment-power-users** (0.70): Power users drive retention
**EV-users-segment-casual** (0.65): Casual users drive growth

**Resolution:** Unresolved. Both valid for different goals.
**Recommendation:** Requires explicit decision (DEC-*)
```

### Step 4: Extract Key Insights

For each theme, formulate 1-3 key insights:

**Insight structure:**
- **Observation:** What the evidence shows
- **Im

Validation Details

Front Matter
Required Fields
Valid Name Format
Valid Description
Has Sections
Allowed Tools
Instruction Length:
4927 chars